/srv/irclogs.ubuntu.com/2009/03/12/#ubuntu-meeting.txt

=== asac_ is now known as asac
=== johnc4511 is now known as johnc4510-laptop
=== ianmcorvidae|alt is now known as ianmcorvidae
=== mib_qtlwb3mk is now known as naymyowin
naymyowinhi07:16
=== mikrox is now known as mikro
=== cjwatson_ is now known as cjwatson
* persia peers about11:58
persiaOK.  Time for the Mobile Team meeting12:00
persia#startmeeting12:00
MootBotMeeting started at 07:00. The chair is persia.12:00
MootBotCommands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE]12:00
persia[LINK] https://wiki.ubuntu.com/MobileTeam/Meeting/2009/2009031212:00
MootBotLINK received:  https://wiki.ubuntu.com/MobileTeam/Meeting/2009/2009031212:00
persia[topic] persia's action items12:00
MootBotNew Topic:  persia's action items12:00
persiaErm.  I'll run through the test cases and put them up tomorrow12:00
persia[topic] NCommander's action items12:01
MootBotNew Topic:  NCommander's action items12:01
* NCommander is trying to pull up the list12:01
* ogra is here now 12:01
ogrado we have lool ? or is he travelling ?12:02
NCommanderI have my jax10 working again, I'll be posting the log today12:02
ogra(i know davidm is)12:02
NCommanderNo progress on anything else due to babbage enablement.12:02
* ogra looks around for StevenK 12:02
persiaNCommander, OK.  DO you want those carried over?12:02
NCommanderpersia, I think the arm-softboot-loader spec probably should me scratched, simply because I'm not going to touch it until karmic unless my TODO flies away and I have nothing left/12:03
NCommanderthe dmesg will be posted today, no need to carry over.12:03
* StevenK shores12:04
persiaI'll carry it anyway, as it's not done now :)12:04
NCommanderfair enough12:04
persiaBut I'll drop the arm-softboot-loader stuff.12:04
ograNCommander, it would still be helpful to have a full researc on the softbootloader stuff in advance to karmic12:04
persia[topic] lool's action items12:04
MootBotNew Topic:  lool's action items12:04
NCommanderogra, I perfectly agree, but I don't think its worth this getting carried over constantly, I'm not touching it until karmic.12:05
persiaIt's on the roadmap anyway.12:05
persiaOK.  I'll carry that over.12:06
persia[topic] ogra's action items12:06
MootBotNew Topic:  ogra's action items12:06
persiaogra, Still three people listed on the Roadmap.  Did the less visible part get done?12:06
ograwell, babbage enablement kept me from doing anytrhing else12:07
persiaWell, since your only action item was "take charge of Babbage enablement", I'm still curious if it got done :)12:07
ograthough i think the babbage image implements a part of selection-of-arm images12:07
ograprobably even the biggest part :)12:08
persiaThat's not a bad thing.12:08
persiaStill, are you the reporting person, or shall this be carried over?12:08
ograhttp://people.ubuntu.com/~ogra/arm/babbage/ has the image ...12:08
MootBotLINK received:  http://people.ubuntu.com/~ogra/arm/babbage/ has the image ...12:08
persiaWe're not there yet :)  That comes later.12:08
ograseems to largely work with a handfull of known bugs12:08
ogra(well, thats my report on babbage enablement ... whgich is an action item ;) )12:09
persiaOK.  I'm claiming you're in charge, and that the action is done then.12:10
persiaMoving on to the roadmap12:10
persia[topic] offline-installer12:10
MootBotNew Topic:  offline-installer12:10
ogranmo progress ... still the rootfs script is our best we have12:11
ogra*no12:11
persia[topic] mobile-setup-wizard12:11
MootBotNew Topic:  mobile-setup-wizard12:11
ograand i'm happy to confirm that bug 328167 isnt caused by the setup the script implements12:11
ubottuLaunchpad bug 328167 in gnome-keyring "gnome-keyring-daemon eating 100% CPU at login in Jaunty" [Medium,Triaged] https://launchpad.net/bugs/32816712:11
persiaOh, good!12:11
ograits actually a g-k-d bug12:11
ogra(since it also shows up on the live image)12:12
ograany news on the setup wizard ?12:13
persiaI'm still trying to figure out exactly how to display the right next steps to complete the procedure with oem-config, but think I'm nearly there.  I ought to be able to report out on a procedure for an OEM install of both mobile flavours for next week.12:14
persiaNote that UNR can be oem installed now, if one doesn't use the launcher (or exposes the Desktop).12:14
persia[topic] arm-library-optimisation12:14
MootBotNew Topic:  arm-library-optimisation12:14
persialool doesn't seem to be around, so we'll skip this one this week12:14
persia[topic] pouslbo-packaging12:15
MootBotNew Topic:  pouslbo-packaging12:15
persiaIn Progress.  The 2D stuff works great.  The 3D stuff doesn't work.  The documentation is lagging.12:15
persia[topic] general-resolution-for-touchscree-handling12:15
MootBotNew Topic:  general-resolution-for-touchscree-handling12:15
ograno progress (sadly)12:15
persia[topic] arm-softboot-loader12:16
MootBotNew Topic:  arm-softboot-loader12:16
ograhttp://www.internettablettalk.com/forums/showthread.php?s=3127852160d9e8ec0680bf3720e08e49&t=25975 the bootmenu at the bootom is very intresting12:16
MootBotLINK received:  http://www.internettablettalk.com/forums/showthread.php?s=3127852160d9e8ec0680bf3720e08e49&t=25975 the bootmenu at the bootom is very intresting12:16
persiaSo, digging into this debate: I agree that we need to get more research and prep work done before karmic starts.12:16
NCommanderno progress w.r.t. to drafting the spec.12:16
persiaNot necessarily any code, but it would be nice to have at least some serious framework for discussion for UDS.12:16
ograright12:17
NCommanderNo, that I agree12:17
ograthe possible existing implementations should be researched (and there are plenty)12:17
persiaOK.  Since we're all in agreement, let's leave this on the roadmap, and keep tracking it.12:17
ograand should be listed in the spec draft12:17
persia[topic] selection-of-arm-images12:17
MootBotNew Topic:  selection-of-arm-images12:17
ograwell ...12:18
ograno gui tzool and i doubt i'll write one12:18
ograbut we have a live image12:18
ogradue to lack of offficially supported kernels it doesnt make real sense to add the additional tool12:18
persiaShould the spec be trimmed, and a follow-up spec be written for a GUI, or will a GUI not be useful in the long-term?12:18
NCommanderNow that the kernel has landed, we just need flash-kernel added, then a d-i port becomes fessible.12:18
ograyeah, spec trimming seems sensible12:19
persia[action] ogra to trim selection-of-arm-images to a smaller scope12:19
MootBotACTION received:  ogra to trim selection-of-arm-images to a smaller scope12:19
ograa gui tool really only makes sense if we have multiple kernels we support12:19
persia[action] ogra to register a new spec for a GUI to assist with kernel addition to arm images12:19
MootBotACTION received:  ogra to register a new spec for a GUI to assist with kernel addition to arm images12:19
persiaRight.  Let's just register that separately, and then discuss again once we have a better picture of the karmic kernel situation.12:20
ograjep12:20
persia[topic] lpia-versus-i38612:20
MootBotNew Topic:  lpia-versus-i38612:20
ograreally depends what platforms we'll support in KK12:20
persialool is away again, so skipping.12:20
persia[topic] mobile-spec-cleanup12:20
MootBotNew Topic:  mobile-spec-cleanup12:20
persiaNo change12:20
persia[topic] bug #29984712:20
MootBotNew Topic:  bug #29984712:20
ubottuLaunchpad bug 299847 in linux "Shared memory operations on very fast ARM hardware suffer from non-atomic operations and race conditions." [High,Triaged] https://launchpad.net/bugs/29984712:20
ogradid you ever recieve the app reviews ?12:21
* ogra knows he didnt do one12:21
persiaThat was mid-application-selection, and no, y'all suck :p12:21
NCommanderWhat app reviews?12:21
NCommanderoh12:21
NCommanderI sent mine!12:21
persiaAnyway, we're talking about a FTBFS bug.12:21
NCommanderAnyway, no real progress. strace has a huge signal to noise ratio due to perl, and I have yet to create  a working test case12:22
ograright, NCommander, any news ?12:22
persiaOK.12:22
persia[topic] bug#32816712:22
* NCommander would like to see if a new kernel this still happens12:22
MootBotNew Topic:  bug#32816712:22
ograupstream acknowledged it as NEW ...12:22
ograno further response yet12:22
persiaCool.  Any likelihood of a solution?12:22
persiaOh well.12:22
ograi saw it working once on the live image and thought it was a clock issue12:22
ograbut that was a red herring12:23
NCommanderAs a datapoint12:23
NCommanderI've never had an issues with g-k-d12:23
ogranow that i have a live image i can work on that to do further debugging though12:23
* NCommander has not been able to reproduce :-/12:23
ograNCommander, it should show up on the live image12:23
=== cjwatson_ is now known as cjwatson
ograif it doesnt for you, please note that in the bug12:23
ograbecause then its somehow HW related12:24
ograi cant actually reproduce it now12:24
ograon all my setups12:24
ograit might as well be kernel related, but we need more data to actually point at a place12:25
NCommanderogra, scratch that, just started up, and its happening for me12:26
ograright12:26
* NCommander hasn't had this issue before expect when running on the liveCD12:26
ograso everyone of us can reproduce it now12:26
persia[topic] bug #28069912:26
MootBotNew Topic:  bug #28069912:26
ubottuLaunchpad bug 280699 in libcanberra "canberra-gtk-play crashed with SIGSEGV in malloc()" [Medium,Invalid] https://launchpad.net/bugs/28069912:26
ograwhich should give us the option to nail it down more easily12:26
NCommanderI think that's the wrong bug number12:27
ograyeah12:27
ogra669 :)12:27
persia[topic] bug #28066912:27
MootBotNew Topic:  bug #28066912:27
ubottuLaunchpad bug 280669 in linux "DMA mode and driver jax10" [Low,Triaged] https://launchpad.net/bugs/28066912:27
persiaSorry.12:27
* ogra once had a car plate with the same issue :) was fun12:27
NCommanderI have my jax10 working again, and I have an lpia image, I'm going to post the dmesg and answer lool's question, hopefully today12:28
persia[topic] bug #33677012:28
MootBotNew Topic:  bug #33677012:28
ubottuLaunchpad bug 336770 in flash-kernel "Problems Installing Jaunty On NSLU2" [Undecided,Confirmed] https://launchpad.net/bugs/33677012:28
ograits gotten a lot further and i think the issues the reporter sees are due to lvm or encryption12:29
ograthought its a bit odd that he hits oom while swap is mounted12:29
ograi'm running a test install today but it will still take several hours12:29
persiaSounds under control then.12:29
ograi think its fixed if you go with all the standards12:30
NCommanderogra, NSLU2 installation requires loads of swap, I've had it OOM with Debian with a 364M swap partition.12:30
persia[topic] bug #33814812:30
MootBotNew Topic:  bug #33814812:30
ubottuLaunchpad bug 338148 in vnc4 "Needs new version from Debian: fails to build with removal of mesa-swx11-source" [High,Triaged] https://launchpad.net/bugs/33814812:30
ograjust selecting special options seems to trigger it12:30
ograthat vnc bug doesnt lok any arm specific12:31
ogra(the rodmap mentionjs it as armel porting specific)12:31
persiaRight.  I'll take a look at it in detail tomorrow, but I agree it doesn't look ARM specific.12:31
ogralooks liek a simple merge bug12:32
ogra*like12:32
persia[action] persia to investigate 33814812:32
MootBotACTION received:  persia to investigate 33814812:32
persia[topic] NSLU2 enablement12:32
MootBotNew Topic:  NSLU2 enablement12:32
ograwell, its tracked on bug #33677012:33
ubottuLaunchpad bug 336770 in flash-kernel "Problems Installing Jaunty On NSLU2" [Undecided,Confirmed] https://launchpad.net/bugs/33677012:33
persiaAnd you said that you think it works if you use the default options, right?12:33
ograas i said above, i assume installing without extra options works12:33
ograright12:33
persiaIf you can confirm that, would you drop it from the roadmap as complete?12:33
ograthe reporter there tries to use encrypted lvm and i assume thats just not possible with the amount of ram12:34
ograbut i need to confirm first12:34
ograyep12:34
persia[topic] Babbage Enablement12:34
MootBotNew Topic:  Babbage Enablement12:34
ografirst manually rolled live image is there12:34
NCommandermorning davidm12:35
davidmNCommander, morning12:35
ograwe're waiting for NCommanders redboot package and for lool's redboot partitioning scheme12:35
=== Igorots is now known as Igorot
ograand a tool to write to the partitions12:35
NCommanderWell, we can build redboot from source now12:35
davidmSorry I'm late and I have to leave again in minutes12:35
NCommander(and its evil toolchain)12:35
ograright, package it :)12:35
NCommanderI have12:35
NCommander(mostly)12:35
NCommanderPackages still need some work, but I can get a working GCC package12:35
* ogra considers it done if a package shows up on jaunty-changes :P12:36
ogra(so dont say "i have" ;) )12:36
NCommanderWell, I packaged it (which I have)12:36
NCommanderI just haven't submitted it :-)12:36
ograif we require a special tool to write to the redboot partitions, this tool should probably go into the package as well12:37
ograplease talk to lool about that12:37
ograit might only use dd with different offests12:37
ograwhich could as well be a flash-kernel addition, but i'm not sure12:37
NCommanderogra, well, at the moment, I'm concerned on building redboot as a package, lool is handling fis/fconfig/flash-kernel AFAIK12:38
ograright12:38
ograthats what i'm saying12:38
ogralool handles the partitioning and will tell us what we need to write to them12:38
ograif there is a special script that might be better going into the redboot package12:39
NCommanderogra, well, at the moment, I only have the EVIL toolchain building12:39
NCommanderand I need to fix up the GCC package so its not just one major blob of files12:39
ograright, do what you need to do12:39
ograthats highest prio12:39
ograwe need redboot packaged to autobuild the images12:40
ograi'm currently blocked on that with liveimage auto-creation12:40
ograa manual image is nice as a first shot, but we need dailies12:40
davidmNCommander, please get the situation solved ASAP, we need to get liveimage auto-creation12:41
davidmI need dailies last week!!!!12:41
davidmNCommander,12:41
ograhaha12:41
davidmNCommander, ^^12:41
ograwithout kernel :P12:41
* NCommander would need a time machine to fix that for you :-/12:41
davidmOK sorry have to run to catch shuttle to office12:41
ograeven the kernel we have atm is very shaky12:41
ograit only works due to several hacks i applied12:42
NCommanderogra, I still can't use a root=USB stick :-/12:42
ogra(in initramfs/casper)12:42
davidmunderstood, but lets get what we can done, done.12:42
ograright12:42
ografirst step is redboot12:42
NCommanderdavidm, its top of my priorities, but full directions on how to build RedBoot are up, so we can generate the blob by hand, but if I'm lucky, we can get a RB binary built, then I need to sitdown and make the packages not be ugly12:42
NCommander(building GCC is extremely nasty ATM)12:42
ograand getting the kernel in shape, which i''m working on with amitk12:42
ogralast step is to make it all work with ubiquity12:43
NCommanderI should have all the initial packaging done by weeks end. Probably take a few more days though to get it all through REVU, through NEW, and into the archive ...12:43
ograsome issues are already on cjwatson's plate for that (partitioning and HW support in d.i for example)12:44
ograNCommander, feel free to ask StevenK, persia and me for REVU reviews ... we should get it through fast12:44
NCommanderogra, oh, I plan to, but I still need to have the packages done to do that :-)12:44
ograindeed12:45
ograpersia, i think thats it for babbage enablement atm12:45
persia[topic] ARM benchmarking12:46
MootBotNew Topic:  ARM benchmarking12:46
NCommanderMostly completed. Two app benchmarks haven't been run, but I have a working LCD again, so I can finish those up once I write the old redboot and a 2.6.26 kernel to my SD card.12:46
ograthats only relevant for karmic, right ?12:47
ograwe wont do any archive rebuild for jaunty12:47
NCommanderogra, right, these have been done awhile, but blocked on my LCD going bing and dying12:47
ograso its not critical to have before beta12:47
ogrababbage enablement is ...12:47
persiaOK.  That's it for the roadmap then.12:48
ograso please priorize babbage above benchmarks12:48
persiaThe only other thing on the agenda is meeting times, so:12:48
NCommanderogra, I know :-)12:48
persia[topic] Discuss moving meeting time to be more convenient for those in UTC-5 through UTC-812:48
MootBotNew Topic:  Discuss moving meeting time to be more convenient for those in UTC-5 through UTC-812:48
ograwell, not enought people around to discuss that imho12:48
persiaI submit that not enough people are around in part because of the current time of the meeting.12:49
ogranot really12:49
* NCommander notes that most of the Americans that are regulars are MIA12:49
persiaAnyone else want to defer to next week?12:49
ograthe US people are all in UK atm12:49
NCommanderogra, I'm not :-P12:49
ogratheoretically it should be a perfect time12:49
ograNCommander, well, but you are here :)12:50
persiaOK.  Deferred.12:50
NCommanderIt should be noted that with the US in daylight savings, the meeting is now an hour lower for us US folks12:50
NCommander(yay for 8am vs 7m)12:50
persiaAny other business?12:50
* ogra hasnt heard lower wrt to time ever :)12:50
ogralooks like we're done ?12:52
persiaOK. Meeting adjourned.12:52
persia#endmeeting12:52
MootBotMeeting finished at 07:52.12:52
persiaI'm feeling lousy today.  If someone else volunteered to write up the minutes, I'd be happy.12:53
* dholbach hugs persia12:53
persiaOh well.  Expect them late then.12:54
ograthanks persia for chairing12:54
persiadholbach, :)12:54
* NCommander hugs persia as well12:54
cjwatsonogra: I'm interested to know what these issues that are on my plate are :-)12:55
ogracjwatson, the partitioning stuff that partman came up with12:55
cjwatsonogra: err, you'll have to remind me12:56
ogracjwatson, and general HW support for babbage, we talked about both before but i'm waiting for a motre finalized kernel12:56
ogra*more12:56
ogracjwatson, be sure i will, its my main work item until beta12:56
ograbut atm i'm focuing mre on getting the images clean than on installability12:57
ogra*focusing more12:57
ogra(one step at a time)12:57
=== MaWaLe1 is now known as MaWaLe
=== mdeslaur_ is now known as mdeslaur
* persia peers about13:57
persiaWho's here for the Java Team Meeting?14:00
ttxo/14:00
persiaRight.  Someone ought popularise these meetings more :)14:01
persiaNothing special on the agenda.14:01
persiarobilad isn't here14:01
persiaslytherin isn't here14:01
persiaThe only remaining obstacle to dropping sun-java5 is that there are some pending SRUs, which I'm hoping to get in process this weekend.  I do expect it can be dropped for beta.14:02
persiattx, How's progrss on the Java-Contents file?14:02
ttxpersia: I haven't make a lot of progress since last week, I was taken on something more urgent.14:03
ttxAs soon as I free one hour or two it should be completed14:03
ttxand the result posted on some people.ubuntu.com location. + code to regenerate it in LP14:04
persiaOK.14:04
persiaHow about plans to attract more Java packagers?14:05
ttxI still have to compile a list of missing libraries that we would target and coordinate with debian-java14:05
ttxso no progress here either, sorry14:06
persiaOK.14:06
persiaAnything else to add or mention?14:06
ttxi had good feedback on my LibraryPackaging doc though14:06
persiaCool!14:06
ttxludovicc and twerner seem to make good progress on the Maven side of things14:06
ttxChecking their work out is on my TODO list... somewhere14:07
persiaExcellent.  You think we'll be in shape for karmic?14:07
ttxWe should import most of what's needed during the debian import window, yes.14:07
ttxI have to look at it deeper to be sure it will be the case14:08
persiaThat should make it easier to adopt more packages, or at least handle those who have signed on to maven completely.14:09
ttxnothing else on my side14:11
persiaI don't have anything else either.14:12
persiaMeeting adjourned then.  Until next week.14:12
ttxThanks persia !14:13
=== kenvandine1 is now known as kenvandine
loolpersia: Thanks for running the mobile meeting15:36
=== Andre_Gondim is now known as Andre_Gondim-afk
=== ian_brasil is now known as ian_brasil_ack
=== ian_brasil_ack is now known as ian_brasil
=== Nicke_ is now known as Nicke

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!